Source note2 min read
A Self-Evolving Agent for Longitudinal Personal Health Management
Source
arXiv
arxiv.org/abs/2607.13940v1
Source
- Source name: arXiv
- Source kind: paper
- Authors / organization: Haoran Li, Jiebi Deng, Tong Jin, Jinghong Han, Yuxin Wang, Zexin Wang, Qingyi Si, Weikang Gong, Xiahai Zhuang, Jia You, Wei Cheng, Jianfeng Feng, Hongcheng Guo
- Venue: arXiv: cs.AI
- Published: 2026-07-15
- Collected: 2026-07-16
- URL: http://arxiv.org/abs/2607.13940v1
- PDF: https://arxiv.org/pdf/2607.13940v1
- DOI: UNKNOWN
Plain-Language Brief
A research paper introduces HealthClaw, a self-evolving agent architecture for personal health management. The system updates support as a person's routines, preferences, measurements, and risks change. It was evaluated on synthetic and biomedical tasks, showing improved accuracy and privacy compared to baselines.
What the Document Says
- The research paper discusses HealthClaw, an open-source agent architecture for personal health management.
- HealthClaw separates shared safety rules and medical knowledge from private longitudinal memory containing profile facts, reusable procedures, and episodic traces.
- After each episode, induction determines what should update the profile, revise a procedure, remain episodic or be excluded.
- The evaluation of HealthClaw involved synthetic year-long benchmark and nine 200-case biomedical tasks.
- Across 900 longitudinal support probes, HealthClaw showed improved answer accuracy compared to current-query prompting, with lower prompt-side context exposure than full-history prompting.
- In 100 privacy probes, HealthClaw produced higher privacy-aware answer quality and fewer unsafe disclosures than baselines.
- Across biomedical tasks, the mean absolute gain in the task-specific primary metric was 27.0 percentage points, with seven gains remaining significant after false-discovery-rate correction.
- The offline benchmarks suggest that HealthClaw supports governed, self-evolving memory for longitudinal personal health agents.
- However, clinical effectiveness requires prospective evaluation.
Key Concepts Mentioned
- HealthClaw
- self-evolving agent
- personal health management
- longitudinal memory
- induction
- synthetic benchmark
- biomedical tasks
- privacy
- clinical effectiveness
Claims Made by the Document
- HealthClaw increases answer accuracy compared to current-query prompting.
- HealthClaw reduces prompt-side context exposure compared to full-history prompting.
- HealthClaw produces higher privacy-aware answer quality and fewer unsafe disclosures than baselines.
- Across biomedical tasks, HealthClaw shows a significant improvement in the task-specific primary metric.
- The offline benchmarks suggest clinical effectiveness of HealthClaw requires prospective evaluation.
Evidence Used
- Results from 900 longitudinal support probes
- Results from 100 privacy probes
- Results across nine 200-case biomedical tasks
- Offline benchmarks
Methods / Approach
UNKNOWN
Limitations Stated by the Authors
UNKNOWN
Limitations Noted by the Agent
Agent interpretation — not from the source.
UNKNOWN
Open Questions
UNKNOWN
Notes for Future Q&A
May include interpretation. Labeled where it does.
UNKNOWN
Raw Abstract / Excerpt
Personal health management unfolds over repeated encounters, yet most health AI systems treat each request in isolation. We developed HealthClaw, an open-source agent architecture that updates support as a person's routines, preferences, measurements and risks change. It separates shared safety rules and medical knowledge from private longitudinal memory containing profile facts, reusable procedures and episodic traces. After each episode, induction determines what should update the profile, revise a procedure, remain episodic or be excluded. We evaluated HealthClaw with a synthetic year-long benchmark and nine 200-case biomedical tasks. Across 900 longitudinal support probes, answer accuracy increased from 0.2% with current-query prompting to 45.7% with HealthClaw, while prompt-side context exposure was 71.7% lower than with full-history prompting. In 100 privacy probes, HealthClaw produced higher privacy-aware answer quality and fewer unsafe disclosures than both baselines. Across the biomedical tasks, the mean absolute gain in the task-specific primary metric was 27.0 percentage points, and seven gains remained significant after false-discovery-rate correction. These offline benchmarks support governed, self-evolving memory for longitudinal personal health agents, although clinical effectiveness requires prospective evaluation. HealthClaw is publicly available at https://github.com/HC-Guo/HealthClaw.
